Designing thermal radiation metamaterials via a hybrid adversarial autoencoder and Bayesian optimization

Zhu, Dan; Guo, Jiaqi; Yu, Gang; Zhao, C.Y.; Wang, Hong; Ju, Shenghong

doi:10.1364/ol.453442

Cited by 8 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bayesian optimization method can be further combined with generative networks. Zhu et al 297 synergistically integrates the capabilities of the adversarial autoencoder and Bayesian optimization techniques for designing narrow-band thermal emitters operating at distinct target wavelengths. The hybrid framework leverages the strengths of both AAE and BO, capitalizing on the efficient exploration of the design space facilitated by AAE and the precision of optimization provided by BO.…”

Section: Thermalmentioning

confidence: 99%

Machine Learning Aided Design and Optimization of Thermal Metamaterials

Zhu,

Bamidele,

Shen

et al. 2024

Chem. Rev.

View full text Add to dashboard Cite

Artificial Intelligence (AI) has advanced material research that were previously intractable, for example, the machine learning (ML) has been able to predict some unprecedented thermal properties. In this review, we first elucidate the methodologies underpinning discriminative and generative models, as well as the paradigm of optimization approaches. Then, we present a series of case studies showcasing the application of machine learning in thermal metamaterial design. Finally, we give a brief discussion on the challenges and opportunities in this fast developing field. In particular, this review provides: (1) Optimization of thermal metamaterials using optimization algorithms to achieve specific target properties. (2) Integration of discriminative models with optimization algorithms to enhance computational efficiency. (3) Generative models for the structural design and optimization of thermal metamaterials.

show abstract

Section: Thermalmentioning

confidence: 99%

Machine Learning Aided Design and Optimization of Thermal Metamaterials

Zhu,

Bamidele,

Shen

et al. 2024

Chem. Rev.

View full text Add to dashboard Cite

show abstract

“…Specifically, the inverse design of optical metasurfaces, [ 10 ] such as high efficiency thermal emitter design, [ 11 ] has been enabled by ML and deep neural network (DNN) models, [ 12 ] including adversarial autoencoders (AAs), [ 13 ] generative adversarial networks (GANs), [ 14 ] and variation autoencoders (VAEs). [ 15 ] Trained ML models suggest optimized structural parameters in a single step to produce a design that exhibits the target optical properties.…”

Section: Introductionmentioning

confidence: 99%

Inverse Design of Photonic Surfaces via High throughput Femtosecond Laser Processing and Tandem Neural Networks

Park,

Grbčić,

Motameni

et al. 2024

Advanced Science

View full text Add to dashboard Cite

This work demonstrates a method to design photonic surfaces by combining femtosecond laser processing with the inverse design capabilities of tandem neural networks that directly link laser fabrication parameters to their resulting textured substrate optical properties. High throughput fabrication and characterization platforms are developed that generate a dataset comprising 35280 unique microtextured surfaces on stainless steel with corresponding measured spectral emissivities. The trained model utilizes the nonlinear one‐to‐many mapping between spectral emissivity and laser parameters. Consequently, it generates predominantly novel designs, which reproduce the full range of spectral emissivities (average root‐mean‐squared‐error < 2.5%) using only a compact region of laser parameter space 25 times smaller than what is represented in the training data. Finally, the inverse design model is experimentally validated on a thermophotovoltaic emitter design application. By synergizing laser‐matter interactions with neural network capabilities, the approach offers insights into accelerating the discovery of photonic surfaces, advancing energy harvesting technologies.

show abstract

“…Material informatics integrated with informatics algorithms for material structure optimization has been demonstrated superior to traditional empirical trialand-error methods in the design of multi-degree-of-freedom thermal functional materials [41,42]. It has high efficiency in thermal transport design [43][44][45], thermoelectric optimization [46][47][48] and thermal radiation design [49][50][51]. Moreover, some optimal structures or devices such as aperiodic GaAs/AlAs superlattice structure with low coherent phonon heat conduction [52] and highly wavelength-selective, multilayer nanocomposite selective thermophotovoltaic emitter [53] have been experimentally fabricated, which demonstrates the applicability and efficiency of the informatics algorithms.…”

Section: Introductionmentioning

confidence: 99%